The invention concerns a method for transmitting speech information, as well as a transmission apparatus and a reception apparatus.
In mobile communication connections for speech transmission, at least one of the parties to the call can be reached only via a radio channel. The speech information is standardly transmitted including the pauses in the conversation that it contains. The transmission of the speech information is exposed to various types of disturbing influences in the radio channel, so that disturbances occur at the receiver side that reduce the transmission quality.
In, for example, digital radio transmissions according to the DECT standard, these disturbances reach bit error rates of 10.sup.-3 to 10.sup.-2 in the maximum range region. These bit errors are distributed uniformly over the actual speaking time and the speech pauses. For the receiving party, the disturbances of the radio transmission manifest themselves as unpleasant interference noise. Standard error correction methods, such as repeated transmission of the disturbed speech segments, are not possible due to the real-time conditions.
In connection with DECT wireless telephones, a method for speech transmission is known from Siemens Components 31, 6 (1993). The transmission in the radio channel ensues with digital signals, whereby the transmission of the speech information also contains speech pauses. At the transmitting apparatus, an A/D conversion of the speech information and a forming of the transmission signals takes place, and an evaluation of the received transmission signals and a D/A conversion takes place at the receiving apparatus. In addition to the speech information, items of control information are also transmitted. The method is correspondingly also transferrable to mobile radiotelephone transmission systems, and is known there. A generalization to analog transmission methods of the speech information is also possible; the A/D conversion and D/A conversion are thereby omitted.
The inventive method makes use of the fact that during a transmission of a two-party call the actual speaking time of an individual party is on average much less than 50% of the duration of the call; thus, the pauses in speech contained in the speech information take up a large part of the transmitted speech information. A correspondingly large portion of the bit errors fall to these pauses in speech.
From the German laid open print DE 42 36 088 A1, it is known to improve speech transmission for wireless telephones according to the DECT standard by temporarily suppressing speech transmission when determined error boundary values are exceeded. However, it is necessary to determine the errors that have occurred, which is possible only at the receiver.
Furthermore, from Blaesner, W.; "Audioprozessor vereinfacht NF-Signalaufbereitung im Mobilfunk," ntz vol. 43, 1990, no. 6, pp. 446-451, an audio processor is known that determines speech pauses at the transmitter and produces a logic signal for switching off the transmission final stage. The transmission energy expended and the background noise can thereby be reduced. However, the disturbances occurring in the transmission channel cannot be corrected during the speech pauses.